This proposal is concerned with measuring the effects of a novel automatic emergency alarm and response system on the psychological and social well being (including the prevention of institutionalization) of elderly persons at risk, including the cost-effectiveness of such an intervention. This is to be accomplished within the framework of a controlled impact study in which elderly persons who are 60 years or older and reside within apartments managed or subsidized by the Boston Public Housing Authority (Boston, Mass.), will be randomized into experimental and control groups and the affects of the intervention evaluated. To be eligible for the program the elderly person must be severely or multiply physically handicapped and clinically judged to be physically vulnerable and/or relatively socially isolated. The major outcome measures to be studied for the effects of use of the automatic emergency alarm and response system are as follows: 1) health status; 2) quality of life indicators (morale, social isolation, etc.); 3) institutionalization; 4) time of death; and 5) the costs of both informal as well as formal health and welfare maintenance services. Finally, since it is hoped that the intervention will have the potential for providing a new service for large numbers of the elderly and disabled nationwide, this study will develop projections of cost effectiveness, guidelines for the system's operation, and assessment procedures for effective client screening.